Sunday , November 18 2018

A Smart and Predictive Heating System Using
Data Fusion Based on the Belief Theory

Améni MAKHLOUF1,2*, Bruno MARHIC1, Laurent DELAHOCHE1, Arnaud CLÉRENTIN1, Hassani MESSAOUD2

1 Laboratoire des Technologies Innovantes (LTI) EA 3899, Dpt. Informatique,
Avenue des Facultés le Bailly, 80000 Amiens Cedex 1, France
makhlouf.ameni@gmail.com

* Corresponding author

2 Laboratoire de Recherche en Automatique,
Traitement du Signal et Image (LARATSI), ENIM,
Rue Ibn El Jazzar, 5000 Monastir, Tunisia

Abstract: This paper investigates the way to model and handle the contextual data uncertainties in order to design a smart heating system that reduces energy consumption. To achieve this, we propose in this paper a data fusion system which provides a trend hypothesis (contextual and near future prediction) with its associated belief. The data to be fused are essentially the occupant’s habits, the weather forecast as well as the notion of thermal comfort associated to the occupant’s activities. Since the data are uncertain, erroneous and heterogeneous, we propose a multilevel data fusion system based on the belief theory of Dempster-Shafer for data combination and the Transferable Belief Model (TBM) for making the decision which will be a challenge. Despite the data complexity, the simulations are very satisfactory in terms of reducing energy consumption.

Keywords: multilevel fusion, dempster-shafer theory, modelling uncertainty, energy saving, building management system.

>>Full text
CITE THIS PAPER AS:
Améni MAKHLOUF*, Bruno MARHIC, Laurent DELAHOCHE, Arnaud CLÉRENTIN, Hassani MESSAOUD,
A Smart and  Predictive Heating System Using Data Fusion Based on the Belief Theory, Studies in Informatics and Control, ISSN 1220-1766, vol. 25(3), pp. 283-292, 2016.

1. Introduction

Buildings are responsible for 40% of energy consumption and 36% of CO2 emissions in the EU. The European Performance of Building Directive has set a target for all new buildings to be near zero-energy consumption by 2020. In France thermal standards RT2012 have been set to reach these directives. The RT2012 [14] aims to decrease the level of CO2 emission for new buildings while the RT2015 aims to optimise and to reduce the energy consumption. The reduction of energy consumption in buildings, called the EE (Energetic Efficiency), depends on a set of elements such as the building’s orientation, the geometry, the insulation, the thermal mass and also of AEE (Active Energetic Efficiency) elements, such as intelligent lighting and of course the conception of a better performing HVAC (Heating, Ventilating, and Air Conditioning) regulation and management. In Northern Europe, developing high performance and intelligent heating systems has become crucial in order to reduce energy consumption [2]. In France, electric heating is the main energy consumer in buildings with 70% represented by residential housing [16]; which justifies the development and the integration of smart-heating regulators.

Many researchers try to find the most adapted solutions for energy management in buildings. Some are focused on energy consumption
reduction in public buildings [3, 8 & 12]. In [13] the authors deal with reducing the daily energy costs in household management without affecting the user’s comfort; they consider the occupant’s expectations as well as the physical constraints, such as energy prices and limitations of the renewable energy power. Other projects [4, 15] have been conducted in this context to evaluate the performance of existing control systems, to test new approaches and their influence on energy consumption as well as on the daily lives of the occupants. In [6, 7] the Artificial Neural Networks (ANN) are used to analyse data from different sensors in order to maintain a stable and comfortable temperature in the building. In [11] the researchers have adapted a neural–fuzzy system for the HVAC system in order to collect internal data dynamically and automatically and regulate the temperature to reach an optimal level of thermal comfort.

A building is a dynamic system; it is in permanent interaction with its internal and external environment. Its modelling depends on internal factors, such as occupation, thermal convection between the different rooms and thermal transfer due to the lighting and electrical appliances, and external ones such as the exterior temperature, the wind speed and direction and the orientation of the room with respect to the sun. These considerations lead us to incorporate the meteorological impact on the building in order to propose a model that is as complete as possible. Weather forecasts have been integrated in the regulation of heating systems based on a predictive control strategy [17] for an automated room (IRA: Integrated Room Automation). The different simulations show that the integration of data from weather forecasts with a stochastic MPC approach provides interesting results in terms of energy savings.

Another significant aspect is the presence of the occupant and the nature of his activity in the building. These are essential factors as they contribute directly to the energy efficiency of the building. The authors in [15] have presented the occupation of the building as a dynamic vector that will be integrated in the cost function of the predictive system. In [21] the authors have presented an interesting method to handle the occupancy of housing. It integrates the fact that the occupancy can be perturbed and deviate from any given planning.

In this paper the reader will see how we tackled and succeeded in reducing a heating system’s energy consumption by processing and fusing heterogeneous numerical data in order to take predictive and sensible decisions whilst maintaining the user’s desired comfort level.

The paper is organized as follows: Section 2 rapidly outlines our contribution. Section 3 briefly introduces the belief theory and the TBM model. In section 4 our approach is detailed. Several simulation case studies are presented and discussed in section 5. Finally in section 6 we discuss the reliability of the proposed system by comparing it to a weighted mean method. Then we conclude the paper.

REFERENCES

  1. CANTIN, R., B. MOUJALLED, G. GUARRACINO, Complexity of Thermal Comfort in Buildings, European Conference of Science (6), 2005.
  2. CDC CLIMATE RESEARCH, Key Figures on Climate France and Worldwide 2015. http://www.cdcclimat.com/IMG//pdf/1410_reperes_2015_eng-hd.pdf, Accessed 05 June 2015.
  3. DALY, D., P. COOPER, P. Z. MA, Understanding the Risks and Uncertainties Introduced by Common Assumptions in Energy Simulations for Australian Commercial Buildings, Energy and Buildings, 75, 2014, 382-393.
  4. DOUNIS, A. I., M. BRUANT, M. J. SANTAMOURIS, G. GUARRANCINO, P. MICHEL, Comparison of Conventional and Fuzzy Control of Indoor Air Quality in Buildings. J. Intell. & Fuz. Syst., 42, 1996, 131-40.
  5. FISCHER, C. Feedback on Household Electricity Consumption: A Tool for Saving Energy. En. Eff., 1, 2008, 79-104.
  6. KALOGIROU, S. A., Applications of Artificial Neural Networks in Energy Systems, A Review, Energy Conversion & Management, 40, 1999, 1073-1087.
  7. KIM, S., J. H. LEE, J. W. MOON, Performance Evaluation of Artificial Neural Network-based Variable Control Logic for Double Skin Enveloped Buildings During the Heating Season, Bld. and Envir., 82, 2014, 328-338.
  8. LAZOS, D., A. SPROUL, M. KAY, Optimization of Energy Management in Commercial Buildings with Weather Forecasting Inputs: A Review, Ren. & Sust. Energy Reviews, 39, 2014, 587-603.
  9. LE, C. A., V.-N. HUYNH, SHIMAZU, A. Y. NAKAMORI, Combining Classifiers for Word Sense Disambiguation based on Dempster–Shafer’s Theory and OWA Operators, D. & K. Eng., 63(2), 381-396.
  10. MARHIC, B., L. DELAHOCHE, C. SOLAU, A.-M. JOLLY-DESODT, V. RICQUEBOURG, An Evidential Approach for Detection of Abnormal Behavior in the Presence of Unreliable Sensors, Inf. Fus., 13, 2012, 146-160.
  11. MARVUGLIA, A., A. MESSINEO, G. NICOLOSI, Coupling a Neural Network Temperature Predictor and a Fuzzy Logic Controller to Perform Thermal Comfort Regulation in an Office Building, Bldg. & Env., 72, 2014, 287-299.
  12. MERCIER, D., G. CRON, T. DENEOUX, M. H. MASSON, Decision Fusion for Postal Address Recognition using Belief Functions, Expert Systems with Applications, 36, 2009, 5643-5653.
  13. MISSAOUI, R., JOUMAA, H., PLOIX, S., BACHA, S., Managing Energy Smart Homes According to Energy Prices: Analysis of a Building Energy Management System, En. & Bldg., 71, 2014, 155-167.
  14. MOLLE, D. P. M. PATRY, RT 2012 et RT existant réglementation thermique et efficacité énergétique. Ed.: Eyrolles, 2011.
  15. MOROŞAN, P.-D., R. BOURDAISA, D. DUMURB, J. BUISSONA, Building Temperature Regulation using A Distributed Model Predictive Control, En. and Buildings, 42, 2010, 1445-1452.
  16. OECD/IEA, Technology Roadmap Energy-Efficient Buildings: Heating and Cooling Equipment, 2011, France.
  17. OLDEWURTEL, F., A. PARISIO, C. N. JONES, D. GYALISTRAS, M. GWERDER, V. STAUCH, B. LEHMANN, M. MORARI, Use of Model Predictive Control and Weather Forecasts for Energy Efficient Building Climate Control, En. & Bldg., 45, 2012, 15-27.
  18. SEITY, Y., P. BROUSSEAU, S. MALARDEL, G. HELLO, P. BÉNARD, F. BOUTTIER, C. LAC, V. MASSON, The Arome-France Convective Scale Operational Model, Monthly Weather Review, 139, 2011, 976-991.
  19. SHAFER, G., A Mathematical Theory of Evidence USA, Princeton Univ., 1976.
  20. SMETS, P., R. KENNES, The Transferable Belief Model, AI, 66, 1994, 191-234.
  21. ZHAO, J., B. LASTERNAS, K. P. LAM, R. YUN, V. LOFTNESS, Occupant Behavior and Schedule Modeling for Building Energy Simulation through Office Appliance Power Consumption Data Mining, En. & Bld., 82, 2014, 341-355.

https://doi.org/10.24846/v25i3y201602